Glitches, Fractional Shares and Differences from NCA Results
Steve Denman reports that he has read my articles on spin-offs, and has entered the data on the Lucent/Avaya reorganization in both bivio and NCA. He reports a difference in the gain on the sale of fractional shares, and wants to know why. He says........

I hope I am not making a pest of myself but this accounting is new to me and I need to be able to answer questions from club members (a group of engineers/computer people).

Please don't worry about making a pest of yourself. I know how disturbing it can be to get two different opinions from professionals to whom you are entrusting your data. The difference arises from the fact that NCA computes the cost for the fractional share sold by taking the average cost of all the spin-off shares received. bivio, if the fractional share can not be identified as coming from a specific lot, uses the FIFO [first-in, first-out] method of selection. Our position is substantiated by the IRS code, regulations, and publications. I will attempt to document this authorization below.

When you acquire a security at two different times for two different prices, you must keep track of the cost of each of your positions [called lots]. When you sell shares of this stock, there are certain rules that must be followed in selecting the lot from which the shares that you are selling came.

I will summarize the IRS code and regulations by saying that, if you can not identify the particular lot from which your stock originated, you must use the FIFO method of selection - in other words, pick the certificates that you first purchased. You absolutely can not use the average cost for all your securities when computing the gain or loss on a sale.

The authority for the above position is contained in the IRS regulation 1.1012. This regulation is summarized in the IRS publication 550, which you can download at

On page 40 of the above publication, it says..................

Identifying stock or bonds sold.
If you can adequately identify the shares of stock or the bonds you sold, their basis is the cost or other basis of the particular shares of stock or bonds.

Identification not possible.
If you buy and sell securities at various times in varying quantities and you cannot adequately identify the shares you sell, the basis of the securities you sell is the basis of the securities you acquired first.

Except for certain mutual fund shares, discussed later, you cannot use the average price per share to figure gain or loss on the sale of the shares.

[emphasis added]

The NCA program uses Average Cost, which, as you can see above, except for mutual funds, is absolutely prohibited by the IRS.

I hope this adequately explains the differences between our figures and those of the NCA program. Your next question might very well be......

What am I to do here? I am trying to keep books in both programs so I can evaluate them. Do I have to keep track of these differences?

I would suggest that you can still make the two programs agree. I would recommend that in the NCA program, you enter nothing in the box for the amount received for fractional shares. Then, in a separate transaction, sell the fractional shares, and in selecting the lots, chose the oldest. This should make the two programs agree, and you should be in compliance with IRS regulations for both.

An additional benefit of doing it that way is that you avoid any possibility of incurring the NCA glitch, talked about in my articles about spin-offs, a glitch that the authors of the NCA program now acknowledge. The glitch comes about in certain circumstances, not all, and results in an erroneous result for the gain/loss on sale of fractional shares, and a corresponding error on the balance sheet. To find out more about these glitches read.........

Lucent/Avaya Spin-off

Ford/Visteon Spin-off

Now, I would be the first to acknowledge that in the general scheme of things, these small differences in fractional sales methods are not large enough to bring the IRS down on our doorsteps. Still, the differences in the two programs are disturbing, and I want to give you the reasons for them. Of course, if we are talking about Berkshire Hathaway, differences in the cost basis and gain on fractional shares can become very material.

I hope this long explanation hasn't been overkill for you. I do want you to have complete confidence in the results supplied by bivio.

Please call on us with any more questions or comments.

Rip West
Saint Paul, MN